Discrimination in Insurance Underwriting Guidelines

Insurance companies use underwriting guidelines to assess risk and determine who qualifies for coverage and at what price. These rules help insurers decide when to assume risk and when to decline coverage based on factors like age, health, driving history, or income. While insurers are legally barred from using race as a factor, they rely on actuarial data to evaluate risk, which can result in legal forms of discrimination.

However, growing scrutiny surrounds what counts as fair vs. unfair discrimination, especially in light of systemic inequality and events like the George Floyd protests in 2020. Critics argue that some underwriting practices such as using ZIP codes or credit scores may serve as proxies for race or socioeconomic status, raising ethical and legal concerns about equity in insurance pricing.

The National Association of Insurance Commissioners (NAIC) sets industry standards and has taken steps to address racial bias in insurance. In response to the George Floyd protests, the NAIC convened a special session to examine how racial discrimination persists, particularly through the use of big data algorithms in underwriting.

While overt race-based discrimination has declined, lawsuits and investigations continue to expose legacy practices like redlining and race-based premiums. These tactics still influence coverage decisions and pricing, especially in auto and life insurance sectors.

Discrimination concerns extend beyond property and casualty insurance. In health insurance, federal rules have drawn criticism such as the 2020 CMS nondiscrimination rule, which was flagged by California’s Insurance Commissioner for potentially limiting access for LGBTQ+ individuals, people with disabilities, and non-English speakers.

Insurance underwriting relies on risk-based discrimination, separating applicants into high- and low-risk categories to set premiums and incentivize safer behavior. This form of discrimination is considered actuarially acceptable, as it helps insurers manage financial exposure.

However, U.S. law prohibits unfair discrimination, which targets protected classes such as race, national origin, sex, or religion. Unfair practices may include charging higher premiums, offering weaker coverage, or denying insurance altogether based on factors unrelated to actual risk. These practices have a long history in the industry and remain under scrutiny by regulators and consumer advocates.

In insurance, disparate impact and unfair discrimination are often confused, but they represent distinct legal concepts. Disparate impact refers to practices that unintentionally harm protected groups such as race or gender without overt bias. It’s a legal framework used to prove discrimination based on outcomes, even when intent isn’t explicit.

Unfair discrimination, by contrast, occurs when individuals with similar risk profiles are treated differently due to factors unrelated to risk. This form of discrimination is explicitly banned in every U.S. state, and regulators closely monitor underwriting practices to ensure compliance.

A 2013 legal review by the University of Michigan Law School found that anti-discrimination laws vary widely across states and insurance types. Alarmingly, many jurisdictions lacked specific protections against race-based unfair discrimination, exposing consumers to unequal treatment. The findings suggest a growing need for stronger federal regulation to ensure consistent enforcement and equity in insurance underwriting nationwide.

Redlining is a historic form of insurance and mortgage discrimination that continues to shape economic inequality today. Originating during the Franklin D. Roosevelt administration, the federal government used underwriting maps to classify neighborhoods by perceived risk. The Home Owners’ Loan Corporation (HOLC) color-coded areas based on housing quality, access to transportation, proximity to industrial sites, and critically racial and ethnic composition.

Neighborhoods with predominantly minority populations were marked in red and labeled “hazardous,” leading lenders and insurers to deny coverage and loans. This practice diverted financial resources from communities of color, stifling homeownership, generational wealth, and access to affordable insurance. Though now illegal, the legacy of redlining still influences underwriting decisions and real estate values.

The 1938 FHA Underwriting Manual explicitly linked neighborhood stability to racial segregation, warning against “incompatible racial and social groups” and labeling integration as a threat to property values. These guidelines, paired with HOLC redlining maps, systematically denied loans and insurance to minority communities, causing long-term damage to real estate values and generational wealth.

Legal reforms followed. The 1948 Supreme Court case Shelley v. Kraemer ruled racial covenants unenforceable, and the Civil Rights Act of 1964 outlawed many forms of racial discrimination including race-based life insurance premiums. The 1968 Fair Housing Act banned redlining by race, while the 1965 HUD Act expanded housing access through grants and subsidies. The 1975 HMDA required lenders to disclose demographic lending data.

Despite these laws, allegations of modern redlining persist, with lawsuits in New York claiming discriminatory practices continued into the 21st century. Critics argue that algorithmic underwriting and ZIP code based pricing still serve as proxies for race.

In life insurance, racial bias was historically entrenched. As Mary L. Heen notes, insurers used racial mortality stereotypes to justify offering Black Americans only two-thirds the benefits of White policyholders. Companies often ignored data like women’s lower mortality rates that contradicted their racial assumptions, revealing that risk wasn’t the true driver of pricing.

Race-Based Premiums in Life Insurance: A Legacy of Discriminatory Underwriting

In 1940, the National Association of Insurance Commissioners (NAIC) published a study on mortality rates by race, which insurers used to justify race-based life insurance premiums. These findings fueled discriminatory underwriting policies that persisted until race-based pricing was formally banned.

During this era, insurers maintained dual rate books charging Black policyholders up to 30% more for “industrial life insurance,” which typically covered only burial costs. These policies offered less coverage at higher prices, reinforcing racial inequities in access to financial protection and generational wealth.

Civil Rights Act of 1964: Ending Legal Race-Based Premiums in Insurance

Race-based insurance premiums remained legal until 1964, when mounting pressure from civil rights activists led to the passage of the Civil Rights Act under President Lyndon B. Johnson. This landmark legislation outlawed many forms of racial discrimination, including underwriting practices that charged higher premiums or offered reduced benefits based solely on race marking a pivotal shift toward equity in the insurance industry.

Auto insurance in the U.S. dates back to 1897, with major reforms emerging in the 20th century. In 1938, New Hampshire introduced assigned risk plans, and by the 1970s, states like South Carolina guaranteed access to auto insurance for eligible drivers. Massachusetts pioneered no-fault insurance in 1970, and later banned the use of protected characteristics in pricing.

Key developments in the 1970s and 1980s included:

• Michigan’s 1977 report calling for oversight of subjective underwriting standards
• Massachusetts’ 1978 system guaranteeing access and banning discriminatory pricing
• Michigan Supreme Court’s 1978 ruling against mandatory no-fault coverage
• GAO’s 1986 study on how pricing restrictions affected insurance cost and availability

Despite these reforms, modern redlining persists. A 2017 investigation by Consumer Reports and ProPublica found disparate auto insurance pricing in states like California and Texas differences that couldn’t be explained by risk alone. Researchers called this a “subtler form of redlining,” where ZIP codes and other proxies may still influence rates unfairly.

Auto Insurance and Algorithmic Bias: Race, Credit Scores, and Legislative Reform

Insurance expert Doug Heller of the Consumer Federation of America (CFA) warns that while auto insurers claim not to ask for race, their pricing algorithms often rely on proxies like ZIP codes and credit scores that disproportionately raise premiums for Black Americans. These practices persist despite growing scrutiny of systemic bias in required coverage.

In 2019, Congress introduced H.R. 3693 and H.R. 1756 to curb the use of income proxies and credit scores in auto insurance pricing. Although neither bill advanced, they spotlighted concerns about algorithmic discrimination and the need for federal oversight.

The use of credit scores in auto insurance began in 1995 with Fair Isaac Corp. (FICO) and ChoicePoint. Critics argue that credit scoring acts as a surrogate for redlining, inflating costs for communities of color without reflecting actual driving risk.

In 2021, Colorado passed a landmark law protecting classes like race, sexual orientation, and gender identity from discrimination in underwriting. The law requires insurers to prove their algorithms don’t result in unfair pricing, ensuring that good drivers aren’t penalized by biased data models. As Chuck Bell of Consumer Reports noted, Colorado now has the tools to enforce fair and equitable auto insurance pricing.

Insurance companies rely on algorithms to calculate rates, manage assets, and streamline underwriting. In 2020, the Federal Housing Administration (FHA) launched its first algorithmic underwriting system for single-family mortgages, aiming to modernize and simplify the process. However, concerns remain about how these systems may reinforce bias in insurance decisions.

Advocates argue that algorithms can amplify discrimination, prompting legislative proposals like the 2020 Data Accountability and Transparency Act. This bill sought to ban the use of personal data to discriminate against protected classes and would have required ongoing bias testing in underwriting models. A revised version, the Diversity and Inclusion Data Accountability and Transparency Act, was introduced in 2021 to further address algorithmic fairness.

At the state level, New York regulators prohibit algorithms that cause disparate impact on protected classes. Yet insurers are barred from collecting protected class data, making it difficult to measure algorithmic effects. Other states including California, Connecticut, Illinois, Maryland, Massachusetts, Michigan, and New Jersey have enacted or considered restrictions on using personal data like genetics, education, employment, and ZIP codes in underwriting.

These developments reflect a growing push for algorithmic transparency and fairness, especially as insurers increasingly rely on data-driven models to set rates and assess risk.

Insurance underwriters assess risk using company-specific guidelines that vary by product type. For life insurance, common criteria include age, gender, health history, marital status, and lifestyle habits like smoking or drinking. Auto insurers, meanwhile, evaluate driving records, age, gender, years of driving experience, and past claims to determine premium rates.

Unfair discrimination occurs when individuals with similar risk profiles are treated differently based on non-risk factors like race. This violates U.S. insurance laws and undermines equitable access to coverage.

What Is Redlining?

Redlining refers to the now-illegal practice of denying loans or insurance based on race or ethnicity. Coined by sociologist John McKnight in the 1960s, the term describes how the Home Owners’ Loan Corporation (HOLC) used color-coded maps to label minority neighborhoods as “hazardous.” These maps led to widespread denial of financial services, fueling the racial wealth gap that persists today.

This article provides a comprehensive look at how race, underwriting practices, and legal frameworks intersect in the insurance industry. While overt discrimination has declined, the continued use of proxies and algorithmic models still drives racial disparities in pricing and access. To ensure fairness, legislative and regulatory reforms especially those targeting algorithmic bias and data transparency are essential for building a more equitable insurance landscape.